elasticsearch中子文檔的子聚合？

Question

說我有一個父文檔

curl -XPUT es-host:9200/index/parent_type/1 -d '{ 

    "foo": "bar" 
}' 

，我有一個子文檔

curl -XPUT es-host:9200/index/child_type/?parent=1 -d '{ 

    "child-foo-1": "child-bar-1", 
    "child-foo-2": "child-bar-2" 
}' 

有沒有什麼辦法讓在父文件聚合子聚合爲我的孩子的文檔數據，有點像，

{ 
    "aggregations": { 

    "foo-aggregations": { 

     "buckets": [ 
     { 
      "key": "bar", 
      "doc_count": 1, 
      "child-foo-1-aggregations": { 

      "key": "child-bar-1", 
      "doc_count": 1 
      }, 
      "child-foo-2-aggregations": { 

      "key": "child-bar-2", 
      "doc_count": 1 
      } 
     } 
     ] 
    } 
    } 
} 

Answer 1

curl -XPOST 'localhost:9200/index/parent/_search?pretty' -d ' 
{ 
    "size": "0", 
    "aggs" : { 
     "name" : { 
      "terms" : { 
       "field": "name.keyword" 
      }, 
       "aggs" : { 
        "child_category": { 
         "children" : { 
          "type": "child_type_name" 
         }, 
         "aggs" : { 
          "data" : { 
           "terms" : { 
            "field": "child_field_text.keyword" 
           } 

          } 
         } 
        } 
       } 
     } 
    } 
}'